Powerful Training Free Inpainting Tool Works for Every SD Model. Official Implementation of "Lanpaint: Training-Free Diffusion Inpainting with Exact and Fast Conditional Inference".

Features

🎨 Zero-Training Inpainting - Works immediately with ANY SD model, even custom models you've trained yourself
🛠️ Simple Integration - Same workflow as standard ComfyUI KSampler
🚀 Quality Enhancements - High quality and seamless inpainting

Example Results

Each example includes:

Original masked image
Full ComfyUI workflow

Quickstart

Install ComfyUI.
Install ComfyUI-Manager
Look up this extension in ComfyUI-Manager. If you are installing manually, clone this repository under ComfyUI/custom_nodes.
Restart ComfyUI.

Installation

Place LanPaint_Nodes.py in your ComfyUI/custom_nodes folder
Restart ComfyUI
Use like regular KSampler with inpainting workflows

Usage

Workflow Setup
Same as default ComfyUI KSampler - simply replace with LanPaint KSampler nodes. The inpainting workflow is the same as the SetLatentNoiseMask inpainting workflow.

Note LanPaint only support binary mask (0,1) with no smoothing. Any mask with smooting will be converted to binary mask during inpainting.

Advanced Options (Optional)

Fine-tune results with these key parameters:

Parameter	Typical Range	Effect
`NumSteps`	1-10	Thinking iterations per step
`Lambda`	4-8	Content preservation strength
`StepSize`	0.05-0.2	Detail refinement intensity

Citation

If you use LanPaint in your research or projects, please cite our work:

@misc{zheng2025lanpainttrainingfreediffusioninpainting,
      title={Lanpaint: Training-Free Diffusion Inpainting with Exact and Fast Conditional Inference}, 
      author={Candi Zheng and Yuan Lan and Yang Wang},
      year={2025},
      eprint={2502.03491},
      archivePrefix={arXiv},
      primaryClass={eess.IV},
      url={https://arxiv.org/abs/2502.03491}, 
}

README.md

LanPaint